Assessment of artery dilation by using image registration based on spatial features

Abstract

The use of affine image registration based on normalized mutual information (NMI) has recently been proposed by Frangi et al. as an automatic method for assessing brachial artery flow mediated dilation (FMD) for the characterization of endothelial function. Even though this method solves many problems of previous approaches, there are still some situations that can lead to misregistration between frames, such as the presence of adjacent vessels due to probe movement, muscle fibres or poor image quality. Despite its widespread use as a registration metric and its promising results, MI is not the panacea and can occasionally fail. Previous work has attempted to include spatial information into the image similarity metric. Among these methods the direct estimation of α-MI through Minimum Euclidean Graphs allows to include spatial information and it seems suitable to tackle the registration problem in vascular images, where well oriented structures corresponding to vessel walls and muscle fibres are present. The purpose of this work is twofold. Firstly, we aim to evaluate the effect of including spatial information in the performance of the method suggested by Frangi et al. by using α-MI of spatial features as similarity metric. Secondly, the application of image registration to long image sequences in which both rigid motion and deformation are present will be used as a benchmark to prove the value of α-MI as a similarity metric, and will also allow us to make a comparative study with respect to NMI.

Entropic One-Class Classifiers

“…Vertices V denote the patterns in S, while edges denote their relations in terms of distance; each edge e ij ∈ E has a weight given by w ij = d 2 (v i , v j ), where d 2 (·, ·) is a suitable (normalized) Euclidean metric. Graph representations are popular for high-dimensional patterns [26,45]. Constructing G allows us also to avoid the computational burden of determining the optimal RS, R; the size of the RS corresponds to the dimensionality of the DS, and therefore usually a proper prototype selection method is used to consider the smallest but most informative subset of R. Since d 2 (·, ·) is Euclidean, G can be used also to estimate the α-order Rényi entropy [14,26] of the underlying data distribution through the computation of the entropic MST (see Sec.…”

“…An edge e ij connecting x i and x j is weighted using a weight based on their distance, |e ij | = d 2 (x i , x j ). The α-order Rényi entropy (1) can be estimated according to a geometric interpretation of an entropic spanning graph of G. Examples of such graphs used in the literature are the MST, k -NN graph, Steiner tree, and TSP graph [9,14,25,36,42,[44][45][46]57]. In this paper, we will focus on the MST [8,44].…”